big problem with sql server 2000 msde sp 4 (stops working by accident)

Hi,
I get the following error-message from sqlserver 2000 (msde edition)
I have no idea what is going wrong.
I was only able to get back to working state by
DBCC CHECKDB ALLOW-DATA-LOSS
(or restoring a backup)
But i do not want the db to crash at my customers ...
In the knowledge-base I found:
"An assertion or Msg 7987 may occur when an operation is performed on an
instance of SQL Server"
without any reason why inconsitencies can occur
This bug is at least known since 5th april 2005 ... and still known for
sql-server 9 (2005 I think)
I hope I oversaw something ... or does SQLServer realy stops to work by
chance?
please help!
2006-07-26 03:43:04.00 spid51 ex_raise2: Exception raised, major=79,
minor=87, severity=22, attempting to create symptom dump
2006-07-26 03:43:04.28 spid51 Using 'dbghelp.dll' version '4.0.5'
*Dump thread - spid = 51, PSS = 0x414491a8, EC = 0x414494d8
Event Type: Error
Event Source: MSSQLSERVER
Event Category: (2)
Event ID: 17052
Date: 7/26/2006
Time: 5:32:26 AM
User: N/A
Computer: HLX02
Description:
Error: 7987, Severity: 22, State: 3
A possible database consistency problem has been detected on database
'faroer'. DBCC CHECKDB and DBCC CHECKCATALOG should be run on database
'faroer'.
For more information, see Help and Support Center at
http://go.microsoft.com/fwlink/events.asp.
Data:
0000: 33 1f 00 00 16 00 00 00 3......
0008: 06 00 00 00 48 00 4c 00 ...H.L.
0010: 58 00 30 00 32 00 00 00 X.0.2...
0018: 07 00 00 00 66 00 61 00 ...f.a.
0020: 72 00 6f 00 65 00 72 00 r.o.e.r.
0028: 00 00 ..Sascha Bohnenkamp wrote:
> Hi,
> I get the following error-message from sqlserver 2000 (msde edition)
> I have no idea what is going wrong.
> I was only able to get back to working state by
> DBCC CHECKDB ALLOW-DATA-LOSS
> (or restoring a backup)
> But i do not want the db to crash at my customers ...
> In the knowledge-base I found:
> "An assertion or Msg 7987 may occur when an operation is performed on an
> instance of SQL Server"
> without any reason why inconsitencies can occur
> This bug is at least known since 5th april 2005 ... and still known for
> sql-server 9 (2005 I think)
> I hope I oversaw something ... or does SQLServer realy stops to work by
> chance?
> please help!
> 2006-07-26 03:43:04.00 spid51 ex_raise2: Exception raised, major=79,
> minor=87, severity=22, attempting to create symptom dump
> 2006-07-26 03:43:04.28 spid51 Using 'dbghelp.dll' version '4.0.5'
> *Dump thread - spid = 51, PSS = 0x414491a8, EC = 0x414494d8
>
> Event Type: Error
> Event Source: MSSQLSERVER
> Event Category: (2)
> Event ID: 17052
> Date: 7/26/2006
> Time: 5:32:26 AM
> User: N/A
> Computer: HLX02
> Description:
> Error: 7987, Severity: 22, State: 3
> A possible database consistency problem has been detected on database
> 'faroer'. DBCC CHECKDB and DBCC CHECKCATALOG should be run on database
> 'faroer'.
> For more information, see Help and Support Center at
> http://go.microsoft.com/fwlink/events.asp.
> Data:
> 0000: 33 1f 00 00 16 00 00 00 3......
> 0008: 06 00 00 00 48 00 4c 00 ...H.L.
> 0010: 58 00 30 00 32 00 00 00 X.0.2...
> 0018: 07 00 00 00 66 00 61 00 ...f.a.
> 0020: 72 00 6f 00 65 00 72 00 r.o.e.r.
> 0028: 00 00 ..
No, SQL Server does not "stop working by accident". Most likely you
have flaky hardware that is causing corruption in your database file.
Tracy McKibben
MCDBA
http://www.realsqlguy.com|||Tracy McKibben schrieb:
> No, SQL Server does not "stop working by accident". Most likely you
> have flaky hardware that is causing corruption in your database file.
well ... are there any chance to find someting to help find the problem
in the log files of the server (or in the dumps) ?
From teh stack dump I cannot see any i/o problems ...
* Short Stack Dump
* 009BA08C Module(sqlservr+005BA08C) (GetOSErrString+00004F68)
* 009BA9B5 Module(sqlservr+005BA9B5) (GetOSErrString+00005891)
* 006EE757 Module(sqlservr+002EE757) (SQLExit+00186C60)
* 005FA686 Module(sqlservr+001FA686) (SQLExit+00092B8F)
* 006BA785 Module(sqlservr+002BA785) (SQLExit+00152C8E)
* 004040DA Module(sqlservr+000040DA)
* 0042ADAE Module(sqlservr+0002ADAE)
* 0042A3D3 Module(sqlservr+0002A3D3)
* 0043638B Module(sqlservr+0003638B)
* 0043288B Module(sqlservr+0003288B)
* 00432542 Module(sqlservr+00032542)
* 00434980 Module(sqlservr+00034980)
* 00432542 Module(sqlservr+00032542)
* 00434980 Module(sqlservr+00034980)
* 00432542 Module(sqlservr+00032542)
* 00861732 Module(sqlservr+00461732) (GetIMallocForMsxml+0006CBB2)
* 0086116B Module(sqlservr+0046116B) (GetIMallocForMsxml+0006C5EB)
* 00434980 Module(sqlservr+00034980)
* 00432542 Module(sqlservr+00032542)
* 005B810A Module(sqlservr+001B810A) (SQLExit+00050613)
* 005B8266 Module(sqlservr+001B8266) (SQLExit+0005076F)
* 004398A5 Module(sqlservr+000398A5)
* 004398A5 Module(sqlservr+000398A5)
* 0041D396 Module(sqlservr+0001D396)
* 00861732 Module(sqlservr+00461732) (GetIMallocForMsxml+0006CBB2)
* 0086116B Module(sqlservr+0046116B) (GetIMallocForMsxml+0006C5EB)
* 005B810A Module(sqlservr+001B810A) (SQLExit+00050613)
* 005B8266 Module(sqlservr+001B8266) (SQLExit+0005076F)
* 00434980 Module(sqlservr+00034980)
* 00432542 Module(sqlservr+00032542)
* 004194B9 Module(sqlservr+000194B9)
* 004193E4 Module(sqlservr+000193E4)
* 00429EAA Module(sqlservr+00029EAA)
* 00415D04 Module(sqlservr+00015D04)
* 00416214 Module(sqlservr+00016214)
* 00415F28 Module(sqlservr+00015F28)
* 0076C7D0 Module(sqlservr+0036C7D0) (SQLExit+00204CD9)
* 007704EF Module(sqlservr+003704EF) (SQLExit+002089F8)
* 0077099C Module(sqlservr+0037099C) (SQLExit+00208EA5)
* 006332EC Module(sqlservr+002332EC) (SQLExit+000CB7F5)
* 0043D005 Module(sqlservr+0003D005)
* 0042598D Module(sqlservr+0002598D)
* 41075309 Module(ums+00005309) (UmsThreadScheduler::ExitUser+00000459)|||Sascha Bohnenkamp wrote:
> Tracy McKibben schrieb:
>> No, SQL Server does not "stop working by accident". Most likely you
>> have flaky hardware that is causing corruption in your database file.
> well ... are there any chance to find someting to help find the problem
> in the log files of the server (or in the dumps) ?
> From teh stack dump I cannot see any i/o problems ...
> * Short Stack Dump
> * 009BA08C Module(sqlservr+005BA08C) (GetOSErrString+00004F68)
> * 009BA9B5 Module(sqlservr+005BA9B5) (GetOSErrString+00005891)
> * 006EE757 Module(sqlservr+002EE757) (SQLExit+00186C60)
> * 005FA686 Module(sqlservr+001FA686) (SQLExit+00092B8F)
> * 006BA785 Module(sqlservr+002BA785) (SQLExit+00152C8E)
> * 004040DA Module(sqlservr+000040DA)
> * 0042ADAE Module(sqlservr+0002ADAE)
> * 0042A3D3 Module(sqlservr+0002A3D3)
> * 0043638B Module(sqlservr+0003638B)
> * 0043288B Module(sqlservr+0003288B)
> * 00432542 Module(sqlservr+00032542)
> * 00434980 Module(sqlservr+00034980)
> * 00432542 Module(sqlservr+00032542)
> * 00434980 Module(sqlservr+00034980)
> * 00432542 Module(sqlservr+00032542)
> * 00861732 Module(sqlservr+00461732) (GetIMallocForMsxml+0006CBB2)
> * 0086116B Module(sqlservr+0046116B) (GetIMallocForMsxml+0006C5EB)
> * 00434980 Module(sqlservr+00034980)
> * 00432542 Module(sqlservr+00032542)
> * 005B810A Module(sqlservr+001B810A) (SQLExit+00050613)
> * 005B8266 Module(sqlservr+001B8266) (SQLExit+0005076F)
> * 004398A5 Module(sqlservr+000398A5)
> * 004398A5 Module(sqlservr+000398A5)
> * 0041D396 Module(sqlservr+0001D396)
> * 00861732 Module(sqlservr+00461732) (GetIMallocForMsxml+0006CBB2)
> * 0086116B Module(sqlservr+0046116B) (GetIMallocForMsxml+0006C5EB)
> * 005B810A Module(sqlservr+001B810A) (SQLExit+00050613)
> * 005B8266 Module(sqlservr+001B8266) (SQLExit+0005076F)
> * 00434980 Module(sqlservr+00034980)
> * 00432542 Module(sqlservr+00032542)
> * 004194B9 Module(sqlservr+000194B9)
> * 004193E4 Module(sqlservr+000193E4)
> * 00429EAA Module(sqlservr+00029EAA)
> * 00415D04 Module(sqlservr+00015D04)
> * 00416214 Module(sqlservr+00016214)
> * 00415F28 Module(sqlservr+00015F28)
> * 0076C7D0 Module(sqlservr+0036C7D0) (SQLExit+00204CD9)
> * 007704EF Module(sqlservr+003704EF) (SQLExit+002089F8)
> * 0077099C Module(sqlservr+0037099C) (SQLExit+00208EA5)
> * 006332EC Module(sqlservr+002332EC) (SQLExit+000CB7F5)
> * 0043D005 Module(sqlservr+0003D005)
> * 0042598D Module(sqlservr+0002598D)
> * 41075309 Module(ums+00005309) (UmsThreadScheduler::ExitUser+00000459)
Doesn't have to be an I/O problem - it could be bad RAM, a bad CPU, any
number of things. You should run some hardware diagnostics on the
machine, perhaps even open an incident with Microsoft - they can help
you decipher the logs.
Tracy McKibben
MCDBA
http://www.realsqlguy.com